Project Summary (30 lines) Less than half of the nine million children in the U.S. with a severe mental disorder ultimately receive care. This under-treatment occurs more among African American than among white children. Alarmingly, when they do receive psychiatric care, African American children appear in the emergency department (ED) 11% more than do whites. This relative reliance on the ED is not justified based on racial differences in mental health prevalence or need, and (for African Americans) portends poorly managed symptoms and increased disease severity into adulthood. Despite this racial disparity in children's psychiatric ED visits, no research examines whether ?supply side? expansions over time in mental health services reduces this disparity. Federally Qualified Health Centers (FQHCs) represent one dynamic aspect of the health system that focus on low-income populations. Psychiatric encounters for African American children seen in FQHCs, moreover, increased by over 31% from 2006-2011. This rapid expansion of care, as well as changes in the supply of mental health professionals, may reduce African American children's reliance on the ED. This issue requires further study given that previous expansions of primary care for low-income populations (e.g., Medicaid in Massachusetts and Oregon) show mixed results?especially in failing to reduce disparities. It is plausible but it remains unknown whether increasing FQHCs can treat previously untreated mentally ill African American children. Our project will merge data on FQHCs, county mental health professionals and all children's psychiatric ED visits (including mood, behavioral disorders, suicide attempts) for ten populous states that report all-payer ED data. We will retrieve county-level data (e.g., minority children served in FQHCs, mental health providers) from the Uniform Data System and other proprietary sources. Data on children's psychiatric ED visits will come from the Health Care Utilization Project. The large dataset (~1.4 million visits) collected across 437 counties and six years (2006-2011) provides substantial place and time variation in FQHC penetration. Importantly, we will also link multiple ED records within child to examine whether patterns of psychiatric ED care (i.e., repeat ?treat and release?) decline following expansions in FQHC and mental health providers. Our rigorous identification strategy will control for state and county-level factors (e.g., Medicaid expansion, Mental Health Parity and Addition Equity Act) that affect help seeking. Results, consistent with goals of NIMH, are expected to advance the knowledge base on how features of FQHCs and the health system affect the delivery of psychiatric services to minorities. Our analyses, moreover, will directly inform health policy given that federal allocation of resources to combat disparities crucially depends on group-level results. Even a precisely estimated null result would amplify calls to focus financing and research on ?demand-side? factors for African-Americans' help-seeking and efforts to coordinate referrals and follow-up care in low-income settings.